Failure Mechanism of Reinforced Concrete Structural Walls With and Without Confinement
Failure Mechanism of Reinforced Concrete Structural Walls With and Without Confinement
Failure Mechanism of Reinforced Concrete Structural Walls With and Without Confinement
A. Benavent-Climent, D. Escolano-Margarit,
University of Granada, Granada, Spain
A. Klenke & S. Pujol
Purdue University, West Lafayette, Indiana, US
1. INTRODUCTION
Structural walls have been commonly used during the past decades as a lateral load resisting system
against earthquake. The field observations during the Maule Earthquake of 2010 in Chile have shown
that due their high bending stiffness structural walls reduced story drifts concentrating the structural
damage in the lower levels. However It is necessary a flexural mode of failure in the plastic hinge
regions to avoid a brittle shear failure and to ensure a desired ductile behaviour. Several authors
studied and proposed the requirements to ensure that the lateral shear strength is higher than the
flexural strength. Krolicki, J. et al (2011) proposed a shear strength model based on previous
experimental research. Hidalgo et al. (2002) studied the behaviour of 26 full-scale shear walls
characterized by shear mode of failure. Pilakoutas & Elnashai (1995) tested nine 1:2.5 scale isolated
shear walls under cyclic loading in order to determine the ductility and the energy absorption potential
of RC walls. Another important factor to take into account in the seismic design of RC structural walls
is the estimation of the inelastic flexural displacement of the structural walls by assuming a
concentration of inelastic curvatures at the wall base which are referred as plastic hinges. For the sake
of simplicity, the inelastic curvatures, ϕi , in a plastic hinge are commonly assumed constant over the
length of the plastic hinge, lp. Once the plastic hinge length is determined the inelastic wall
displacement, ∆i , can be easily obtained by integrating the curvatures over the wall height. An
accurate assessment of the plastic hinge length is essential to obtain reliable estimations of the
inelastic displacements. Researchers during the last decades have proposed several expressions in
order to estimate the plastic hinge length in RC columns and beams (Bae, S. & Bayrak, O. (2008);
Priestly, M. J. N., & Park, R. (1987); Sheikh, S. A., & Khoury, S. S. (1993)). However there is a lack
of information in the case of structural walls. Dazio, A. et al (2009) suggested several strain limits for
plastic hinge analysis based on the curvature measured at the base of six shear walls during a cyclic
loading test. Bohl, A., & Perry, A. (2011) proposed a plastic hinge length based on the results of non-
linear finite element analyses.
This paper presents the results from two large-scale reinforced concrete structural wall tests that were
conducted at Purdue University. One of the specimens differs from the other in the inclusion of
confinement reinforcement at the boundary elements. The main goal of this paper is to discuss the
effects of confinement on failure mode. The experimental results of unit curvature and unit strain
distribution presented in this paper also contribute to a better understanding of the plastic hinge length
in structural walls.
2. EXPERIMENTAL INVESTIGATION
Two large-scale reinforced concrete structural walls were tested under quasi-static cyclic loading
conducted at Purdue University. One of the specimens called W-MC-C herein after had confinement
reinforcement while the other called W-MC-N herein after did not have any confinement
reinforcement.
Specimen W-MC-C shown in Figure 1 was designed to meet ACI-318-11 ACI Committee 318 (2011)
confinement reinforcement requirements. The longitudinal reinforcement was comprised of 4 #8 bars
for the confinement element and 3 #4 bars in the wall. The confinement reinforcement is comprised of
#2 stirrups around the boundary elements spaced at 2.5 in c. to c. up to 5 ft.. Also there are #3 ties that
are spaced at 5 in. c. to c. spacing in the wall. Specimen W-MC-N shown in Figure 1 has the same
layout except for the confinement reinforcement. Both specimens had mechanical couplers at the base
of the wall which torque the longitudinal bars from the wall and the footing together. The yield stress,
fy, and ultimate stress, fu , of the reinforcement as well as the compressive stress, fck , tensile stress, fct ,
and Moodule of elasticity, Ec , of the concrete can be found in Table 1.
The structural walls were fixed to the strong floor by 8 large diameter post-tensioning bars. An
auxiliary loading structure was connected on top of the wall to two symmetrical hydraulic actuators
with hinges at both ends. The force capacity of the actuators was 110 kips and a displacement capacity
of 30 inch each. The axial load was applied by two steel beams, placed on the wall head perpendicular
to the loading direction. The load was applied by four post-tensioning bars connected to both ends of
the axial load beams and directly to the strong floor. The bars were post-tensioned by 4 jacks situated
on the live anchor at the top of the walls. Two load cells named LC1 and LC2 in Figure were placed
between the jack and the axial load system to control and to keep the axial load constant throughout
the tests. Two auxiliary steel beams were placed on either side of the wall parallel to the loading
direction to prevent the out of plane failure of the walls. Figure shows the experimental set up during
the tests.
During each cycle the response of the walls, load and displacements, was measured with the following
sensors: (i) Load cells LC1 and LC2 measured the applied load. (ii) Load cells LC3 and LC4 measured
the axial load. (iii) Displacement encoders, ENC 1 to 8 and 12, measured the displacements in the
direction of the loading while displacement encoders ENC 9 to 11 measured the out of plane
displacements. (iv)The coordinates of 62 non contact coordinate-tracking targets forming a 1 by 1 ft
grid were measured at the maximum drift instant of each cycle. The notation and position of
instruments are shown in Figure .
Figure 4: Instrumentation
Initially the axial load was applied with a value of 0.1 fc’Ag. The axial load was kept approximately
constant during the tests. The imposed horizontal displacement history at the walls head is shown in
0.12% 0.25% 0.50% 0.75% 1.0% 1.5% 2.0% 2.5% 3.0%
4
Top displacement (inch)
0 Time
-2
-4
Figure . As can be observed the cyclic displacement consisted in applying three cycles at each drift
level. The drift levels considered were 0.12, 0.25, 0.5, 0.75, 1.0, 1.50, 2, 2.5 and 3.0% expressed as
percentage of the wall height.
0.12% 0.25% 0.50% 0.75% 1.0% 1.5% 2.0% 2.5% 3.0%
4
0 Time
-2
-4
Figure 5: Loading history
4. TESTS RESULTS.
First cracking was observed in cycle #1 at a load of 46 kips and 0.17 inch of displacement. First yield
occurred in cycle #10 at a load of 137 kips and 0.98 inch of displacement. Compression vertical cracks
and spalling at the base of the boundary elements was observed from cycle #11 in advanced. The
maximum strength of 163 kips was reached at cycle #25 at 3.93 inch of displacement. The wall
collapsed in cycle #26 due to the buckling of the boundary element reinforcement. Figure a show the
load versus top displacement hysteresis curve. As can be seen the behaviour of the wall was clearly
controlled by a flexural mode of failure. Figure show photographs of the crack patterns at the base for
several limit states. The maximum crack width measured during the tests was 0.005, 0.03 and 0.15
inch for the flexural cracking, first yield and ultimate capacity limit states.
First cracking was observed in cycle #1 at a load of 42 kips and 0.16 inch of displacement.
Compression vertical cracks and spalling at the base of the boundary elements was observed from
cycle #9 in advanced. First yield occurred in cycle #10 at a load of 137 kips and 0.98 inch of
displacement. The maximum strength of 155 kips was reached at cycle #14 at 2.62 inch of
displacement. The wall collapsed in cycle #19 due to the buckling of the reinforcement. Figure b show
the load versus top displacement hysteresis curve. As can be seen the behaviour of the wall was clearly
controlled by a flexural mode of failure. Figure show photographs of the crack patterns at the base for
several limit states. The maximum crack width measured during the tests was 0.005, 0.025 and 0.15
inch for the flexural cracking, first yield and ultimate capacity limit states.
(a) (b) (c)
Figure 7: Crack patterns for limit states: (a) Flexural cracking. (b) First yield (c) Ultimate capacity
30 30
Load (kips)
0 0
-30 -30
-60 -60
-90 -90
-120 -120
-150 -150
-180 -180
-5 -4 -3 -2 -1 0 1 2 3 4 5 -5 -4 -3 -2 -1 0 1 2 3 4 5
Top displacement (inch) Top displacement (inch)
(a) (b)
Figure 8: Load versus top displacement hysteresis curve (a) W-MC-C (b) W-MC-N
Table 2 summarizes the values of displacement, ∆, and force, V, for the following limit states
proposed by Priestly et al. (2007) i) flexural cracking ∆cr and Vcr ii) First yield ∆y and Vy iii) Ultimate
capacity ∆u and Vu. Comparing the results between specimens its worth noting that there was not a
great difference in the response for the flexural cracking and first yield states. For the ultimate
capacity state, although a similar ultimate lateral load was reached by both specimens, the one with
confinement had a greater lateral deformation, hence a greater ductility as can be observed in Figure .
Load (kips)
30 30
0 0
-30 -30
-60 -60
-90 -90
-120 -120
-150 -150
-180 -180
-4 -3 -2 -1 0 1 2 3 4 -5 -4 -3 -2 -1 0 1 2 3 4 5
Top displacement ∆ (inch) Top displacement (inch)
(a) (b)
Figure 9: Load versus top displacement skeleton curve. (a) W-MC-C (b) W-MC-N
The maximum curvature for the different drift tests was obtained using the normal strains measured by
the non-contact sensors grid overall the wall. Figure shows the distribution along the wall height of
the mean values of the unit curvatures obtained for the 3 cycles at each drift level. As can be seen de
maximum curvature observed at the same drift was almost equal for both specimens. The curvature
profile seems to be approximately linear within a height equal to 0.5hw. Considering the plastic hinge
length as the height above the foundation for which the curvature was equal to the yield curvature the
results presented herein suggest that the plastic hinge length is approximately also 0.5 hw.
110 110
-0.12% 0.12% -0.12% 0.12%
100 -0.25% 0.25% 100 -0.25% 0.25%
-0.51% 0.51% -0.5% 0.50%
-0.74% 0.74% -0.75%
90 90 -1.5%
0.75%
-1% 1% 1.5%
-1.48% 1.48% -2%
80 80 2%
-1.98% 1.98%
-2.47% 2.47%
70
Wall Height (inch)
70
Wall Heigt h(inch)
-2.97% 2.97%
60 60
50 50
40 40
30 30
20 20
10 10
0 0
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 -1.0 -0.5 0.0 0.5 1.0
Curvat ure (1/inch)x103 Curvat ure (1/ inch)x103
(a) (b)
Figure 10: Curvature distribution over the height at different drift sets. (a) W-MC-C (b) W-MC-N
Based on the measurements from a dense array of optical sensors the normal strains were determined
along the wall height. Figure 2 show the strains measured in a 60inch x 60inch grid at the wall base for
the 0.75% and 2% drift tests. As can be seen for the 0.75% drift test the unit tensile strains are greater
for the W-MC-C wall with maximum unit strain value of approximately ε=12000 µm/m versus the
ε=9000 µm/m measured for the W-MC-N. On the other hand the maximum unit compressive strain
was oppositely with values approximately of ε=-4000 µm/m versus the ε=6000 µm/m measured for the
W-MC-N specimen. It is also worth noting that the depth of the neutral axis is greater for the W-MC-
N specimen. In the case of the 2.0% drift test similar unit tensile strains with maximum values
approximately of ε=30000 µm/m were measured. Despite the fact that both similar ranged similar
maximum tensile strains the specimen W-MC-C concentrated the inelastic strains in a region bellow
0.5 hw while in the W-MC-N were spread over that range.
60 60
W-MC-C Drif t 0.75% W-MC-N Drif t 0.75%
-15000
0 -14000
-13000
48 48 -12000
-11000
-10000
0
-9000
36 36 -8000
-7000
-6000
-5000
24 24 -4000
0 -3000
-2000
-1000
0
12 12 1000
2000
Wall height (inch)
3000
4000
0 0 5000
0 12 24 36 48 60 0 12 24 36 48 60 6000
7000
60 60
W-MC-C Drif t 2% 8000
W-MC-N Drif t 2% 9000
10000
11000
48 48 0 12000
13000
14000
0 15000
36 36 16000
17000
18000
19000
24 24 0 20000
21000
22000
23000
24000
12 12 25000
26000
27000
28000
0 0 29000
0 12 24 36 48 60 0 12 24 36 48 60 30000
6. CONCLUSIONS.
This paper presented the results of cyclic loading tests on two large-scale reinforced concrete
structural walls conducted at Purdue University. One of the structural walls had confinement
reinforcement meeting ACI-318-11 requirements while the other wall did not have any confinement
reinforcement. The objective in this research was to investigate the effects of confinement on failure
mode. The main findings of this experimental study may be summarized as follows:
• The confinement did not affect to the yielding force nor to the maximum lateral load.
• The inclusion of confinement reinforcement increased the maximum lateral displacement
capacity, hence the ductility and energy dissipation.
• The inelastic curvatures are concentrated at the wall base, approximately at a height of 0.5
times the wall depth.
• Specimen W-MC-C concentrated the inelastic strains in a region below 0.5 hw while in the W-
MC-N were spread over that range.
AKCNOWLEDGEMENT
Financial support from the Spanish Ministry of Science and Innovation, Project BIA2008/00050 and Fonds
Européen de Dévelopment Regional (FEDER) is acknowledged.
REFERENCES
ACI Committee 318 (2011). 318-11: Building Code Requirements for Structural Concrete and Commentary.
American Concrete Institute.
Bae, S. & Bayrak, O. (2008). Plastic hinge length of reinforced concrete columns. ACI Structural Journal,
105:3, 290-30.
Bohl, A., & Perry, A. (2011). Plastic hinge lengths in high-rise concrete shear walls. ACI Structural Journal,
108:2, 148-157.
Dazio, A., Beyer, K., & Bachmann, H. (2009). Quasi-static cyclic tests and plastic hinge analysis of RC
structural walls. Engineering Structures. 31:7, 1556-1571. doi:10.1016/j.engstruct.2009.02.018
Hidalgo, P. A., Ledezma, C. A., & Jordan, R. M. (2002). Seismic Behavior of Squat Reinforced Concrete Shear
Walls. Earthquake Spectra, 18:2, 287. doi:10.1193/1.1490353
Krolicki, J., Maffei, J., & Calvi, G. M. (2011). Shear Strength of Reinforced Concrete Walls Subjected to Cyclic
Loading. Journal of Earthquake Engineering, 15(sup1), 30-71. doi:10.1080/13632469.2011.562049
Pilakoutas, K., & Elnashai, A. (1995). Cyclic Behaviour of Reinforced Concrete Cantilever Walls, PartI:
Experimental Results. ACI Structural Journal, 92:3, 271-281.
Pilakoutas, K., & Elnashai, A. S. (1995). Cyclic Behavior of Reinforced Concrete Cantilever Walls , Part II :
Discussions and Theoretical Comparisons. ACI Structural Journal. 92:4, 425-433.
Priestly, M.J.N., Calvi, G.M. & Kowalsky, M.J. (2007). Displacement-Based Seismic Design of Structures.
IUSS Press, Pavia, Italy.
Priestly, M.J.N. & Paulay, T. (1992). Seismic Design of Reinforced Concrete and Mansory Buildings. John
Wiley & sons, Inc.
Priestly, M. J. N., & Park, R. (1987). Strength and Ductility of Concrete Bridge Columns Under Seismic
Loading. Aci Structural Journal, January-February, 61-76.
Sheikh, S. A., & Khoury, S. S. (1993). Confined concrete columns with stubs. ACI Structural Journal, 90, 414–
414.